From Univers

United Kingdo

Foundation Tr

Centre for Ca

Kingdom (D.S

Royal Free Lo

Drs Knight, Q

Taylor is fund

ried out with s

Accuracy and Reproducibility of Right VentricularQuantification in Patients with Pressure and Volume

Overload Using Single-Beat Three-DimensionalEchocardiography

Daniel S. Knight, BSc (Hons), MBBS, MRCP, Agata E. Grasso, MD,Michael A. Quail, MB ChB (Hons), MRCPCH, Vivek Muthurangu, MD, MRCPCH,

Andrew M. Taylor, MD, FRCR, FRCP, Christos Toumpanakis, MD, PhD, Martyn E. Caplin, BSc, MD, FRCP,J. Gerry Coghlan, MD, FRCP, and Joseph Davar, FRCP, MD, PhD, London, United Kingdom

Background: The right ventricle is a complex structure that is challenging to quantify by two-dimensional (2D)echocardiography. Unlike disk summation three-dimensional (3D) echocardiography (3DE), single-beat 3DEcan acquire large volumes at high volume rates in one cardiac cycle, avoiding stitching artifacts or longbreath-holds. The aim of this study was to assess the accuracy and test-retest reproducibility of single-beat 3DE for quantifying right ventricular (RV) volumes in adult populations of acquired RV pressure or volumeoverload, namely, pulmonary hypertension (PH) and carcinoid heart disease, respectively. Three-dimensionaland 2D echocardiographic indices were also compared for identifying RV dysfunction in PH.

Methods: A prospective cross-sectional study was performed in 100 individuals who underwent 2D echocardi-ography, 3DE, and cardiac magnetic resonance imaging: 49 patients with PH, 20 with carcinoid heart disease,11with metastatic carcinoid tumorswithout cardiac involvement, and 20 healthy volunteers. Two operators per-formed test-retest acquisition and postprocessing for inter- and intraobserver reproducibility in 20 subjects.

Results:: RV single-beat 3DE was attainable in 96% of cases, with mean volume rates of 32 to 45 volumes/sec.Bland-Altman analysis of all subjects (presented asmean bias6 95% limits of agreement) revealed good agree-ment for end-diastolic volume (�2.36 27.4mL) and end-systolic volume (5.26 19.0 mL) measured by 3DE andcardiac magnetic resonance imaging, with a tendency to underestimate stroke volume (�7.5 6 23.6 mL) andejection fraction (�4.66 13.8%) by 3DE. Subgroup analysis demonstrated a greater bias for volumetric under-estimation, particularly in healthy volunteers (end-diastolic volume, �11.9 6 18.0 mL; stroke volume,�11.2 6 20.2 mL). Receiver operating characteristic curve analysis showed that 3DE-derived ejection fractionwas significantly superior to 2D echocardiographic parameters for identifying RV dysfunction in PH (sensitivity,94%; specificity, 88%; area under the curve, 0.95; P = .031). There was significant interobserver test-retest biasfor RV volume underestimation (end-diastolic volume, �12.5 6 28.1 mL; stroke volume, �10.6 6 23.2 mL).

Conclusions: Single-beat 3DE is feasible and clinically applicable for volumetric quantification in acquired RVpressure or volume overload. It has improved limits of agreement compared with previous disk summation 3Dechocardiographic studies and has incremental value over standard 2D echocardiographicmeasures for iden-tifying RV dysfunction. Despite the ability to obtain and postprocess a full-volume 3D echocardiographic RVdata set, the quality of the raw data did influence the accuracy of the data obtained. The technique performsbetter with dilated rather than nondilated RV cavities, with a learning curve that might affect the test-retestreproducibility for serial RV studies. (J Am Soc Echocardiogr 2015;28:363-74.)

Keywords: Three-dimensional echocardiography, Magnetic resonance imaging, Right ventricular function,Pulmonary hypertension, Carcinoid heart disease

ity College London Medical School, Royal Free Campus, London,

m (D.S.K.); the Department of Cardiology, Royal Free London NHS

ust, London, United Kingdom (D.S.K., A.E.G., J.G.C., J.D.); the UCL

rdiovascular Imaging, University College London, London, United

.K., M.A.Q., V.M., A.M.T.); and the Neuroendocrine Tumour Unit,

ndon NHS Foundation Trust, London, United Kingdom (C.T., M.E.C.).

uail, and Muthurangu are funded by the British Heart Foundation. Dr

ed by the National Institute for Health Research. This work was car-

upport of the GOSH/ICH NIHR Biomedical Research Centre.

Reprint requests: Daniel S. Knight, BSc (Hons), MBBS, MRCP, University College

London, Royal Free Campus, Division of Medicine, Rowland Hill Street, London,

NW3 2PF, United Kingdom (E-mail: danielknight@doctors.org.uk).

0894-7317/$36.00

Copyright 2015 by the American Society of Echocardiography. Published by

Elsevier Inc. This is an open access article under the CC BY license (http://

creativecommons.org/licenses/by/3.0/).

http://dx.doi.org/10.1016/j.echo.2014.10.012

363

Abbreviations

CMRI = Cardiac magneticresonance imaging

COV = Coefficient of variation

EDV = End-diastolic volume

EF = Ejection fraction

ESV = End-systolic volume

ICC = Intraclass correlation

coefficient

PH = Pulmonary hypertension

ROC = Receiver operating

characteristic

RV = Right ventricular

RVOT = Right ventricular

outflow tract

SV = Stroke volume

TAPSE = Tricuspid annular

plane systolic excursion

3D = Three-dimensional

3DE = Three-dimensional

echocardiography

2D = Two-dimensional

2DE = Two-dimensionalechocardiography

364 Knight et al Journal of the American Society of EchocardiographyMarch 2015

Quantification of right ventricular(RV) size and function is prog-nostic in congenital and acquiredheart disease.1-4 The mostconvenient imaging modality forassessing the right ventricle istwo-dimensional (2D) echocardi-ography (2DE). However, thisis limited by the crescentic RVchamber shape and complex ge-ometry, with inflow and outflowportions in different planes.5,6

Thus, cardiac magnetic resonanceimaging (CMRI) has become thegold-standard imaging modalityfor RV quantification.7

Unfortunately,CMRI is expensive,time consuming, and of limitedavailability compared with echo-cardiography.

One possibility to overcomethe limitations of 2DE is three-dimensional (3D) echocardiog-raphy (3DE), compared againstCMRI in a range of congenitaland acquired diseases for RVvolumetric quantification.8

Three-dimensional echocar-diography traditionally uses thedisk summation method toreconstruct the right ventricle af-ter sequential slice acquisitionover consecutive electrocardio-

graphically gated heartbeats.9 This technique, however, is limited bybreath holding throughout successive cardiac cycles, stitching artifactsduring acquisition, and difficulties identifying inlet and outflow re-gions in the basal slices during postprocessing.10

More recently, ultrasound transducer technology allows the real-time acquisition of a 90� � 90� data set in a single cardiac cycle.11

We therefore compared RV volumetric quantification by single-beatfull-volume 3DE against CMRI in homogenous patient populationsof acquired RV pressure and volume overload, namely, pulmonaryhypertension (PH) and carcinoid heart disease, respectively. We alsosought to determine the potential incremental value of 3DE versus2DE in PH and to evaluate the test-retest reproducibility of 3DE forboth the acquisition and postprocessing components.

METHODS

Study Population

We performed a prospective cross-sectional study that enrolled 100participants in sinus rhythm with no contraindications to magneticresonance imaging, all of whom underwent comprehensive 2DE,single-beat 3DE of the right ventricle, and CMRI within 2 hours ofone another. The participants were divided into four subgroups:

� A group of 49 consecutive patients with PH (diagnosed by right heart cath-eterization as a mean pulmonary artery pressure >25 mmHg and a pulmo-nary capillary wedge pressure <15 mm Hg12) who presented for diagnosisand/or follow-up of PH by clinical evaluation and/or right heart catheteriza-

tion as a disease model of RV pressure overload. The etiologies of PHincluded idiopathic (n = 9), connective tissue disease associated (n = 32),and chronic thromboembolic disease (n = 8). Exclusion criteria comprisedclinically significant restrictive or obstructive lung disease identified by pul-monary function tests, arrhythmia, and known independent left-sidedcardiac disease unrelated to PH.

� A group of 20 consecutive patients undergoing 2DE for diagnosis and/orfollow-up of carcinoid heart disease13 were studied as a disease model ofRV volume overload.

� A control group of 20 healthy volunteers affiliated with our institution whowere age and sex matched to the PH group.

� A control group of 11 age- and sex-matched patients with metastatic neuro-endocrine tumor whowere screened as negative for carcinoid valvular heartdisease.

All control participants were eligible for study inclusion if they hadno cardiac symptomatology, had no medical histories of cardiacdisease including hypertension, and were not taking any cardiac med-ications. Normal 2D transthoracic echocardiographic findings werealso required to exclude any occult structural cardiac disease beforestudy inclusion.

The institutional research ethics committee approved the study,and informedwritten consent was obtained from all patients and con-trol subjects.

2DE

All patients underwent comprehensive 2D andDoppler transthoracicechocardiography in the left lateral decubitus position using theAcuson Siemens SC2000 cardiac ultrasound system (SiemensHealthcare, Erlangen, Germany), with a 4V1c transducer (frequencybandwidth, 1.25–4.5 MHz). A standard study protocol was followedin conjunction with American Society of Echocardiography guide-lines for chamber quantification14,15 and the British Society ofEchocardiography guidelines for PH assessment16 as appropriate.RV function was assessed using M-mode tricuspid annular plane sys-tolic excursion (TAPSE); RV fractional area change, calculated as[(end-diastolic area) � (end-systolic area)/end-diastolic area] � 100;and mean RV free wall peak systolic strain using syngo VectorVelocity Imaging (Siemens Medical Solutions USA, Inc, MountainView, CA). A three-beat 2D echocardiographic digital clip of an apicalfour-chamber view optimized for RV visualization was acquired andexported to Velocity Vector Imaging, and 10 to 15 endocardial pointswere plotted in end-systole from the lateral to the medial tricuspidannulus.17 The adequacy of speckle-tracking was visually checkedand manually adjusted as required.

3DE

Image Acquisition. Single-beat full-volume 3D echocardiographicRV data sets were acquired using the 4Z1c matrix-array transducer(frequency bandwidth, 1.5–3.5 MHz; maximum depth, 30 cm;maximum field of view, 90� � 90�). Probe position started fromthe apical four-chamber view with the patient in the left lateral decu-bitus position. Both the patient and transducer positions were subse-quently modified for optimal simultaneous visualization of thetricuspid valve, cardiac apex, infundibulum, and RV outflow tract(RVOT) as assessed by the real-time 2D four-chamber, basal sagittal,and coronal views, and by inclusion of the RV chamber in the pyrami-dal data set. In our experience, a more lateral apical windowwith pos-terior tilt of the probe tail was beneficial to visualize the infundibulumand RVOT in the coronal window. Image depth and sector widthwere adjusted for maximal visualization of the right ventricle at the

Figure 1 Example of the stepwise process of RV reconstruction by 3DE. (A) The cavity area is optimized in the three orthogonal views,and landmarks are identified (mitral valve indicated by the blue asterisk, tricuspid valve indicated by the red asterisk). Note the RVOTseen in the coronal view (indicated by the purple asterisk). (B)Having traced endocardial borders in three orthogonal views, the semi-automated border tracking results are displayed for inspection. Note how the purple guideline (indicated by the purple asterisk) bi-sects the tricuspid valve and RVOT in the short-axis views. This corresponds to the coronal RV reconstruction (highlighted), withclearly delineated RV inflow and outflow portions. A4C, Apical four-chamber.

Journal of the American Society of EchocardiographyVolume 28 Number 3

Knight et al 365

highest volume rate. At least three 3D echocardiographic RV data setswere acquired during a breath-hold to ensure optimal image quality,which was subjectively graded on a five-point scale ranging from zero(very poor) to four (perfect).18 A score of two or less was attributed ifultrasound dropout was evident in greater than half of the RVOTborder.

Postprocessing. Full-volume 3D echocardiographic RV data setswere imported into the on-cart RV Analysis application. Manualadjustment of the RV data set was initially required to ensure the

correct orientation of four-chamber, sagittal, and coronal slices; maxi-mize the RV cavity area and identify themost apical RV view on visualassessment of the four-chamber window; and allow the identificationof cardiac landmarks. This process was performed in a stepwiseapproach by rotation and angulation of the four-chamber window,with manipulation of this plane causing the simultaneous adjustmentof the other two (sagittal and coronal) orthogonal planes (Figure 1A).Both atrioventricular valves followed by the left ventricular apex wereidentified as anatomic landmarks. When the apex of a dilated rightventricle overrode that of the left ventricle, the most apical cardiac

366 Knight et al Journal of the American Society of EchocardiographyMarch 2015

point was identified with the left ventricular apex marker. End-diastolic and end-systolic frames were assigned by visual identificationof the largest and smallest RV four-chamber areas, respectively.

Endocardial RV borders were traced at end-diastole andend-systole in four-chamber, basal sagittal, and coronal views. Thesoftware algorithm obliges the operator to intersect the endocardialborder tracing in sequential views with crosshair reference markersthat are positioned in response to endocardial border traces from apreceding view. Therefore, correction of a previous slice tracing wasundertaken when a crosshair position suggested a prior tracing error.Trabeculae were included in the blood pool volume. To assist withRVOT delineation in the basal sagittal view, the insertion point ofthe RV myocardium at the interventricular septum was routinelyincluded in the endocardial tracing.

At the final stage, the algorithm presents the results of semiauto-mated contour tracking for the four-chamber, coronal and basal,middle and apical short-axis views. Misalignment of endocardial con-tours prompted identification of the region of suboptimal trackingfollowed by manual correction of the original tracing. Automatedvolumetric reconstruction was accepted only once the semiauto-mated endocardial border tracking was visually satisfactory and repre-sented meaningful RV shapes in all views (Figure 1B), as optimizationof this final reconstruction stage significantly affects the results gener-ated.19 The algorithm from which the final RV volume is generatedhas been previously described.20,21

Test-Retest Reproducibility of 3DE. Reproducibility was stud-ied in 20 randomly selected subjects (14 with PH, one with carcinoidheart disease, and five healthy volunteers) for both the 3D echocar-diographic acquisition and postprocessing stages by two independentsonographers (D.S.K. and A.E.G.), as described previously.22 The twosonographers had equal experience with 2DE but differing levels ofexperience with 3D echocardiographic RV full-volume acquisition(10 and 3 months, respectively). Sonographer 1 (D.S.K.) obtained a3D echocardiographic RV data set, after which sonographer 2(A.E.G.) independently obtained a 3D echocardiographic RV dataset. Then, sonographer 1 acquired a second separate 3D echocardio-graphic RV data set. The sonographers, who were blinded to eachother’s results, performed postprocessing of their own 3D echocar-diographic RV data sets. Data sets for intraobserver test-retest repro-ducibility were postprocessed separately at time intervals of >2weeks.

Cardiac MRI

Image Acquisition. All cardiac magnetic resonance images wereacquired using a 1.5-T magnetic resonance scanner (Avanto;Siemens Healthcare) using a 12-element phased-array coil for signalreception and the body coil for signal transmission. A vector electro-cardiographic systemwas used for cardiac gating. Ventricular volumesand great vessel flow were measured in all patients. Volumetric RVdata were obtained using either retrospectively gated balancedsteady-state free precession (n = 19) cine imaging of contiguousshort-axis slices23 or real-time radial k-t sensitivity encoding imaging(n = 81) of contiguous transaxial slices24 depending on the pathologyunder investigation and the patient’s ability to hold his or her breath.Real-time radial k-t sensitivity imaging allows the collection of highspatiotemporal resolution real-time images during free breathingand is part of the standard clinical CMRI work flow at our institutionin the pediatric PH population.25 Transaxial RV slices were preferredfor the PH cohort and their respective control population because of

the relative preservation of longitudinal versus radial RV function thatis manifest in this condition.26 Blood flow data were acquired in theascending aorta, in the right and left branch pulmonary arteries, andat the level of the atrioventricular valves using a velocity-encoded pro-spectively triggered spiral phase-contrast magnetic resonance flowsequence.27 This provided an internal check for the RV volumetricdata.

Postprocessing. All image postprocessing was performed using‘‘in-house’’ plug-ins for the open-source OsiriX Digital Imaging andCommunications in Medicine software.24,28,29 Endocardial RVborders were traced manually at end-diastole and end-systole, thetime points of which were identified by the largest and smallest RVcavity areas, respectively. The inclusion of RV trabeculae was thesame as that performed by 3D echocardiographic postprocessing.Ventricular stroke volume (SV) was the difference between end-diastolic volume (EDV) and end-systolic volume (ESV), and ejectionfraction (EF) was calculated as (SV/EDV)� 100. Phase-contrast mag-netic resonance flow data were segmented using a semiautomaticvessel edge detection algorithm with manual operator correction.28

The CMRI data sets for the patients who underwent 3D echocardio-graphic test-retest reproducibility scans were also tested for interob-server (D.S.K. and M.A.Q.) and intraobserver postprocessingreproducibility.

Statistical Analysis

Statistical analysis was performed using SPSS version 21.0 (IBMCorporation, Armonk, NY) and Prism 6.0b for Mac (GraphPadSoftware, Inc, La Jolla, CA). Normally distributed continuous datawere expressed as mean6 SD. Systematic differences between mea-surements were evaluated with Student paired t tests (two tailed),with Pearson correlation coefficients used to assess the relationshipbetween 3DE- and CMRI-derived RV volumes and EF. Differencesbetween the four participant subgroups were analyzed using one-way analysis of variance, with the Tukey post hoc tests identifyingwhich specific means differed. P values < .05 were considered statis-tically significant. Image scoring data were nonparametrically distrib-uted, represented by medians with 25th and 75th percentiles. Ranksum tests were used for comparisons of image scoring data, withtheMann-WhitneyU test and the Kruskal-Wallis test used for compar-isons of two and three independent groups, respectively.

Intermodality, interobserver, and intraobserver agreement wasstudied using the Bland-Altman method,30 whereby the mean differ-ence was presented as the bias and 95% limits of agreement aroundthe bias expressed as themean difference6 1.96 SDs. Differences be-tween test-retest measurements were analyzed by one-way repeatedmeasures analysis of variance, with the Bonferroni post hoc test iden-tifying which specific means differed. The Greenhouse-Geissercorrection was used if the assumption of sphericity had been violated.Test-retest variability was expressed using intraclass correlation coeffi-cients (ICC), relative differences, and coefficients of variation (COVs).The ICC was quantified by the two-way random-effects model withabsolute agreement. An ICC > 0.85 was considered excellent.Relative differences were calculated by taking the absolute differencebetween two observations divided by the mean of the repeated ob-servations and expressed as a percentage. COVs were calculated asthe standard deviation of the difference between two acquisitionsdivided by their mean value and expressed as a percentage.31 ACOV # 10% was considered excellent.

Table 1 Clinical characteristics of study populations

Variable

PH

(n = 46)

Carcinoid heart

disease (n = 19)

Healthy

volunteers

(n = 20)

Carcinoid (no

valvulopathy) (n = 11) P*

Age (y) 56 6 13 63 6 8 50 6 12 59 6 10

Women 35 (76%) 7 (37%) 15 (75%) 7 (64%)

Height (cm) 164 6 9 171 6 10 169 6 8 168 6 10 .035

Weight (kg) 69 6 17 72 6 18 72 6 12 77 6 20 .54

Body surface area (m2) 1.8 6 0.2 1.8 6 0.3 1.8 6 0.2 1.9 6 0.3 .37

Heart rate (beats/min) 74 6 14 67 6 13 68 6 9 69 6 12 .19

Mean PASP on RHC (mm Hg) 44 6 16

Pulmonary vasodilators

Endothelin antagonist 21 (46%)

PDE-5 antagonist 31 (67%)

Prostanoid infusion 2 (4%)

Oral prostanoid 1 (2%)

Prostaglandin receptor agonist 1 (2%)

Carcinoid heart disease: affected valves TV = 19 (100%), PV = 13 (68%), MV = 3 (16%), AV = 3 (16%)

EDV (mL/m2) 87 6 26 100 6 35 64 6 14 52 6 8 <.0001

ESV (mL/m2) 52 6 25 33 6 15 22 6 7 16 6 5 <.0001

EF (%) 43 6 14 68 6 7 65 6 7 71 6 7 <.0001

3D echocardiographic temporalresolution (volumes/sec)

34 6 5 32 6 7 40 6 5 45 6 6 <.0001

AV, Aortic valve; ESV, end-systolic volume;MV, mitral valve; PASP, pulmonary artery systolic pressure; PDE-5, phosphodiesterase-5; PV, pulmo-nary valve; RHC, right heart catheterization; TV, tricuspid valve.

Data are expressed as mean 6 SD or as number (percentage).

*One-way analysis between groups.

Journal of the American Society of EchocardiographyVolume 28 Number 3

Knight et al 367

Receiver operating characteristic (ROC) curves were derived for2D and 3D echocardiographic parameters to identify CMRI-derived RV EFs of <50% in patients with PH and healthy volun-teers.32 Patients with carcinoid disease were excluded from thisanalysis to avoid the confounding effects of severe valvular regurgita-tion on ventricular function. The area under the ROC curve for anechocardiographic parameter is presented together with the optimalcutoff threshold for detecting CMRI-derived RV EF < 50%, definedas the value of the parameter that corresponded to the highest sumof sensitivity and specificity. The Delong method was used tocompare the areas under the curve between ROC curves33

(Analyse-it Software, Ltd, Leeds, United Kingdom).

RESULTS

Population Characteristics and 3DE Technical Data

Of 100 individuals who were recruited, four had unobtainable RVechocardiographic windows. The clinical characteristics and 3D echo-cardiographic technical data of the final cohort of 96 subjects arepresented in Table 1. Patients with PH had significantly larger andimpaired right ventricles than controls, whereas the right ventriclesof patients with carcinoid heart disease were also significantly dilatedbut with preserved EFs. The dilated right ventricles of the patientgroups resulted in a significantly lower mean volume rate compared

with controls because of the greater 3D sector angles (P < .001), butthe median image quality score was significantly higher amongpatients (3.00; interquartile range, 2.00–3.00) than controls (2.00 in-terquartile range, 1.00–3.00) (P < .001). The image quality amongthree successive, equally populated subgroups of patients significantlyimproved with increasing experience with 3DE (Figure 2; P = .031).There was a trend, albeit not statistically significant, for greater differ-ences in SV between modalities with worse subjective image scores(Figure 3; P< .13 for percentage intermodality difference in SV for im-age score groups 1 and 2 combined vs groups 3 and 4 combined).

Volumetric Analysis by 3DE versus CMRI

Correlation coefficients showed good to excellent correlations be-tween modalities for RV metrics in patient groups and moderate togood correlations for control subjects (Table 2). RV volumes andEFs by 3DE showed differences with CMRI in both patient groups,with a bias for underestimating SVand EF but with overall acceptablelimits of agreement (Figure 4). By contrast, 3DE underestimated EDVfor control subjects (Table 3), with a consequent negative bias forquantifying SV in this group (Figure 5).

RV Quantification by 3DE and 2DE versus CMRI

Three-dimensional echocardiographic EF was the most superiorechocardiographic parameter for identifying CMRI-derived RV EF

Figure 3 Box-and-whisker plots of differences in stroke volumebetweenmodalities (expressedasapercentageof theCMRI refer-ence value) for image scoring groups 1 to 4. There was a trend,albeit not statistically significant, for the intermodality differenceto increasewith reductions in subjective image score.Median per-centage intermodality differences in stroke volumeby image scoregroup were as follows: group 1, 22% (interquartile range [IQR],�1% to 26%); group 2, 11% (IQR, �2% to 23%); group 3, 9%(IQR, 3% to 16%); group 4, 2% (IQR, �6% to 19%).

Table 2 RV volumes and EFs by single-beat full-volume 3DEversus CMRI

Group Measurement 3DE CMRI P*

PH EDV (mL) 158 6 53 154 6 52 .043ESV (mL) 100 6 44 92 6 47 <.0001SV (mL) 58 6 18 63 6 17 .011EF (%) 39 6 11 43 6 14 .00029

Carcinoid heart

disease

EDV (mL) 182 6 69 185 6 71 .21ESV (mL) 67 6 28 62 6 3 .01SV (mL) 115 6 42 124 6 45 .014EF (%) 64 6 5 68 6 7 .001

Healthy

volunteers

EDV (mL) 105 6 26 117 6 27 <.0001ESV (mL) 41 6 12 41 6 14 .80SV (mL) 65 6 16 76 6 18 <.0001EF (%) 61 6 5 65 6 7 .014

Carcinoid (no

valvulopathy)

EDV (mL) 88 6 21 98 6 27 .05ESV (mL) 32 6 13 30 6 14 .24SV (mL) 56 6 10 68 6 14 .009EF (%) 64 6 7 71 6 7 .004

ESV, End-systolic volume.Data are expressed as mean 6 SD.

*Paired Student t tests.

Figure 2 Box-and-whisker plots of subjective image qualityscores among three successive subgroups of patients (group1 acquired in the earliest phase of the study, group 3 in the latestphase of the study). Image quality significantly improved withincreasing experience with 3DE (P = .031).

368 Knight et al Journal of the American Society of EchocardiographyMarch 2015

< 50% (Figure 6; P = .031), with sensitivity of 94%. A fractional areachange of 39% (sensitivity, 85%) was the best conventional 2D echo-cardiographic measure, superior to both peak systolic strain andTAPSE (P = .0443). TAPSE was the weakest marker to predictCMRI-derived RV EF < 50%, with sensitivity of 56% at a cutoffthreshold of 19 mm.

Test-Retest Intraobserver and InterobserverReproducibility

Limits of agreement were acceptable for intra- and interobserver 3Dechocardiographic studies, with good to excellent ICCs (Table 4).However, there was a significant interobserver bias for underestimat-ing RV EDV (P= .001; Table 5) that resulted in underestimation of SV(P = .002) and EF (P = .033), with accompanying large interobserverCOVs and relative differences. Moreover, despite no significantdifferences between intraobserver EDV and ESV, the differencestranslated into statistically significant test-retest differences for SV(P = .032) and EF (P = .005). The interobserver and intraobserverreproducibility for RV volumes and EF by CMRI showed no signifi-cant bias and superior limits of agreement compared with 3DE.

DISCUSSION

This study demonstrates the feasibility of single-beat full-volume 3DEfor RV quantification in, to our knowledge, the largest homogenousacquired RV pressure- and volume-overloaded patient populations.Single-beat 3DE is an agreeable technique compared with CMRI,albeit with significant differences especially in subjects with nondi-lated right ventricles. Furthermore, 3D echocardiographic parametersare of incremental benefit for RV functional quantification comparedwith traditional 2DE measures.

Accurate quantitation of RV size and function is important in manycongenital and acquired cardiac diseases and is of particular relevancein our study populations. RV size and function are of greater prog-nostic significance in PH than the afterload to which the right heartis exposed,34,35 with RV EF being the key determinant of outcomeregardless of changes in pulmonary vascular resistance afforded bypulmonary vasodilator therapy.36 Similarly, right heart dilatation isindependently associated with poor outcomes in patients withadvanced carcinoid heart disease.37 However, the right ventricle re-sponds differently to pressure- and volume-overload conditions,with dilatation occurring in both but with relative preservation of

Figure 4 Bland-Altman analysis of bias (black solid line) and 95% limits of agreement (red dashed line) for 3DE versus CMRI quan-tification of RV EDV, ESV, SV, and EF in patients with PH and carcinoid heart disease.

Table 3 Bias, limits of agreement, and correlation between single-beat full-volume 3DE and CMRI for RV volumes and EFs

Group Measurements Bias 6 SD Limits of agreement r P*

All subjects EDV (mL) �2.3 6 13.7 �29.1 to 24.5 0.97 <.0001ESV (mL) 5.2 6 9.5 �13.4 to 23.9 0.98 <.0001SV (mL) �7.5 6 11.8 �30.6 to 15.7 0.94 <.0001EF (%) �4.6 6 6.9 �18.2 to 9.0 0.91 <.0001

PH EDV (mL) 4.0 6 13.1 �21.6 to 29.7 0.97 <.0001ESV (mL) 8.4 6 10.6 �12. 3 to 29.1 0.98 <.0001SV (mL) �4.3 6 10.8 �25.5 to 17.0 0.82 <.0001EF (%) �4.8 6 8.3 �21.1 to 11.5 0.81 <.0001

Carcinoid heart disease EDV (mL) �3.1 6 10.1 �22.9 to 16.8 0.99 <.0001ESV (mL) 5.4 6 8.2 �10.6 to 21.4 0.96 <.0001SV (mL) �8.6 6 13.9 �35.9 to 18.6 0.95 <.0001EF (%) �3.8 6 4.1 �11.9 to 4.2 0.82 <.0001

Healthy volunteers EDV (mL) �11.9 6 9.0 �29.5 to 5.8 0.94 <.0001ESV (mL) �0.4 6 6.7 �13.6 to 12.9 0.88 <.0001SV (mL) �11.2 6 10.1 �31.0 to 8.7 0.84 <.0001EF (%) �3.9 6 6.5 �16.6 to 8.8 0.51 .021

Carcinoid (no

valvulopathy)

EDV (mL) �10.1 6 15.0 �39.6 to 19.4 0.84 .001ESV (mL) 2.1 6 5.5 �8.7 to 12.9 0.92 <.0001SV (mL) �12.2 6 12.3 �36.3 to 11.9 0.53 .096EF (%) �6.2 6 5.6 �17.1 to 4.7 0.69 .019

ESV, End-systolic volume.

*Pearson correlation coefficient.

Journal of the American Society of EchocardiographyVolume 28 Number 3

Knight et al 369

function in elevated preload rather than afterload. What remains un-clear is to what extent this preserved EF represents normality of func-tion in the presence of severe tricuspid regurgitation, a valvular lesioncommon to all patients in our carcinoid heart disease cohort.

The incremental benefit of 3DE over 2DE has previously beenshown in congenital heart disease,38 and single-beat 3DE showedsimilar added value over 2DE metrics in acquired RV pressure over-

load. Although this is due in part to equivalent parameters being as-sessed by 3DE and CMRI, it is, importantly, also a reflection of thelimitations of conventional 2D echocardiographic measures. TAPSEhad the poorest sensitivity for detecting lowRVEF in PH, with a cutoffof 19 mm having the highest combined sensitivity and specificity. Thisis higher than the recommended threshold of 16mm for detecting RVdysfunction,14 suggesting that TAPSE would have performed worse

Figure 5 Bland-Altman analysis of bias (black solid line) and 95% limits of agreement (red dashed line) for 3DE versus CMRI quan-tification of RV EDV, ESV, SV, and EF for subjects in the control populations.

Figure 6 ROC curves for 3DE, fractional area change (FAC), RVfree wall peak systolic strain by speckle-tracking echocardiog-raphy, and TAPSE to identify RV dysfunction (defined as RVEF < 50% on CMRI).

370 Knight et al Journal of the American Society of EchocardiographyMarch 2015

by current guidelines in our cohort. The rocking motion of the rightventricle in pressure overload can give rise to apparently normalTAPSE values,39 and TAPSE also does not account for the radialcomponent of RV function that contributes significantly to RV EF.26

By contrast, fractional area change was the most superior 2D echocar-

diographic marker for identifying RV dysfunction in PH, most likely areflection of being the only 2D echocardiographic marker that ac-counts for radial function. These findings are consistent with previousstudies comparing 2D echocardiographic markers of RV function inPH40 and suggest that 3DE may have an important additive role inassessing RV function.

RV quantification by echocardiography is advantageous throughbeing more readily available and less expensive than CMRI. Sincethe first use of 3DE for RV volumetric quantification,18 improvementsin matrix-array transducer technology permit the simultaneous visual-ization of orthogonal 2D RV planes at the time of acquisition. Thetechnique used in this study allows a pyramidal data set of up to90� � 90� to be acquired at higher temporal resolutions than previ-ously reported for 3DE.41 Acquisition of a full volume in a singleheartbeat avoids stitching artifacts associated with acquiring slicesover serial heartbeats and also confers the advantage of shorterbreath-hold durations. These reasons might explain the narrowerlimits of agreement for RV volumetric parameters between single-beat 3DE and CMRI compared with previous data from adult PHgroups using the disk summation method.41,42

The disadvantages of echocardiography include constraints thatafford inadequate transthoracic windows, including body habitus,hyperinflated lungs, and chest deformities. Acquisition and postpro-cessing was feasible in 96% of subjects, consistent with previously re-ported studies using the technique.43,44 However, patients withsignificant lung disease were excluded to ensure that PH was thepredominant disease process in the RV pressure-overload group,and this may in turn have biased the echogenicity of the study popu-lation. Although all postprocessed 3D echocardiographic data setshad a reconstructed RV polygon that tracked throughout the cardiaccycle, 45% of studies were judged by subjective image scoring to havesome endocardial dropout of the outflow portion of the rightventricle. This was reflected by a trend toward increasing differences

Table 4 Interobserver and intraobserver reproducibility for RV volumes and EF by 3DE and CMRI

Variable EDV (mL) ESV (mL) SV (mL) EF (%)

3DE intraobserver

ICC 0.992 0.974 0.96 0.906

COV (%) 3.0 6.6 8.0 6.9

RD (%) 4.3 9.4 11.3 9.8

Bias �0.2 4.6 �4.7 �3.6

LOA �16.2 to 15.8 �12.8 to 22.0 �19.0 to 9.7 �12.2 to 5.0

SD 8.2 8.9 7.3 4.4

CMRI intraobserver

Bias �2.6 �2.4 �0.1 0.7

LOA �15.4 to 10.2 �11.3 to 6.5 �11.8 to 11.6 �5.9 to 7.2

SD 6.5 4.6 6.0 3.4

3DE interobserver

ICC 0.955 0.965 0.867 0.827

COV (%) 7.7 8.0 16.6 9.4

RD (%) 10.3 11.4 23.5 13.3

Bias �12.5 �2.0 �10.6 �4.0

LOA �40.0 to 15.1 �24.0 to 20.1 �33.2 to 12.1 �16.2 to 8.3

SD 14.1 11.3 11.6 6.3

CMRI interobserver

Bias �1.9 �2.80 1.1 0.9

LOA �18.2 to 14.4 �13.1 to 7.5 �9.3 to 11.5 �4.1 to 5.8

SD 8.3 5.2 5.3 2.5

ESV, End-systolic volume; LOA, limits of agreement; RD, relative difference.

Table 5 Interobserver and intraobserver test-retest RV metrics by 3DE

Sonographer 1 Sonographer 2

Variable First (S1.1) Second (S1.2) P* S1.1 vs S1.2 Acquisition P* vs S1.1 P* vs S1.2

EDV (mL) 145 6 63 145 6 62 NS 133 6 59 .003 .003

ESV (mL) 78 6 44 83 6 42 NS 76 6 39 NS NS

SV (mL) 67 6 31 63 6 29 .032 57 6 27 .002 .046

EF (%) 48 6 13 44 6 12 .005 44 6 11 .033 NS

ESV, End-systolic volume.

Data are expressed as mean 6 SD.

*One-way repeated-measures analysis of variance with Bonferroni post hoc test.

Journal of the American Society of EchocardiographyVolume 28 Number 3

Knight et al 371

in SVs between modalities with decreasing image quality, with a me-dian difference of $11% when the RVOT was incompletely visual-ized. This is a consistent problem with 3DE that has been welldocumented previously and is due to the anterior position of the rightventricle in the thorax. Postprocessing software extrapolates theendocardial borders during semiautomated border tracking,45 andhence although it is possible to analyze data sets with incompleteRVOT visualization, the accuracy of reconstructions will most likelydeteriorate with progressive dropout in the outflow tract.

When comparing studies of RV quantification by 3DE, the homo-geneity of the study population must be taken into account. Our pop-ulations of acquired RV disease were favorable for the 3Dechocardiographic postprocessing software algorithm, because it is

set up for an adult-shaped right ventricle rather than a congenitalheart.46 This may be a reason why our limits of agreement were nar-rower than reported in congenital heart disease.45 No substantial biaswas observed in either the PH or carcinoid heart disease group, butsubgroup analysis showed that EDVs, and consequently SVs, wereunderestimated in controls. This is despite the higher temporal reso-lution of images in this group and is likely a result of low spatial reso-lution with single-beat 3DE. Lower spatial resolution confers lessability to resolve myocardium and trabeculae, thus directing the oper-ator to trace the endocardium further inside the RV cavity and henceunderestimate volumes. This is supported by previous data showinggreater variability and negative bias for 3DE to quantify RV volumesin nondilated right hearts.47 Conversely, RVendocardial delineation is

372 Knight et al Journal of the American Society of EchocardiographyMarch 2015

known to be easier in the setting of RV hypertrophy or dilatation forboth magnetic resonance imaging and 3DE41,47 and is reflected bythe higher image quality scores observed with our disease cohorts.

The progressive increase in 3D echocardiographic image qualityover the study duration reflects a significant learning curve with thetechnique, as also described in previous studies.38 This is importantclinically, as follow-up studies will vary depending on operator expe-rience for both acquisition and postprocessing. Few studies so far haveaddressed 3D echocardiographic test-retest reproducibility for boththe acquisition and postprocessing stages.22,41 Our interobservertest-retest study demonstrated a second operator bias for EDV under-estimation, conferring lower SV and EF measurements. This was asystematic error likely reflecting relative operator inexperience withthe technique. The susceptibility of 3DE to underestimate RV vol-umes has been well documented,8 and our data suggest that operatorexperience is related to this underestimation.

Furthermore, nonsignificant differences in intraobserver EDV andESV nevertheless resulted in significant differences in SV and EFwhen the errors in the raw volumes were combined. Given thatsmall changes in endocardial border delineation are known toconfer significant changes in 3DE-derived volumetric parameters inthe left ventricle,48 this is also likely to be a problem with 3DE recon-struction of the right ventricle too. This is clinically important becausea change of as little as 10 mL in SV by CMRI is clinically significant inPH,49 but a change of this magnitude may bemasked by 3DE’s repro-ducibility error and/or the degradation of accuracy found with poorerquality 3D echocardiographic data sets. For example, the interob-server measurement of RV SV by 3DE showed a significant biaswith a standard deviation more than double that of CMRI. OurCMRI reproducibility data show narrow limits of agreement, withno major bias between observers, consistent with previous reproduc-ibility studies of RV quantification by transaxial slices50,51 andsensitive enough to detect small changes in RV indices on serialstudies.

Limitations

This study was a single-center study based on acquisitions made byone sonographer with experience using single-beat 3DE for RVvolumetric quantification. As demonstrated by our test-retest repro-ducibility, results cannot be applied across operators with variableexperience in 3D echocardiographic RV analysis. Patients with ar-rhythmias were excluded because of the extra variability introducedby irregular cardiac cycles when comparing modalities. Single-beatacquisition is advantageous over traditional disk summationtechniques that are limited by stitching artifacts due to irregular R-Rintervals, and this patient group requires further investigation withthe technique.

We found a bias to underestimate cavity size, particularly in controlsubjects, but the study was not designed to compare the accuracy ofthe technique in patients with large versus small cavity sizes. Thisquestion should be addressed in a separate prospectively designedanalysis of large versus small right ventricles. The sensitivity andspecificity values for 3DE and 2DE to identify CMRI-derived RVEF< 50% were calculated by applying the ROC cutoff values to thesame patients used to derive them as described previously,38 hencerepresenting a ‘‘best case’’ scenario. A more appropriate methodwould be to identify cut-off values using ROC analysis in a derivationgroup, then prospectively evaluate the cutoff values in a separate testgroup in whom outcomes could be verified independently. Thus the

diagnostic performance of the cutoff values found in this study needsto be confirmed independently.

Finally, the study was not designed to provide CMRI test-retestreproducibility similar to the 3D echocardiographic study design foracquisition and postprocessing. However, CMRI does not have thesame acquisition window restrictions inherent to transthoracic echo-cardiography, as contiguous transaxial RV slices of fixed thickness areacquired from the base of the right heart to the main pulmonaryartery with the patient in the supine position. However, this differencein technique methodology is a potential source of discrepancy, withthe reference standard of CMRI building volumes frommultiple slicescompared with the full-volume data sets of 3DE.10

CONCLUSIONS

Single-beat full-volume 3DE is a feasible technique for quantifying RVsize and function in acquired right heart pressure and volumeoverload. The limits of agreement of 3DE are acceptable comparedwith CMRI but may not be sensitive enough to detect small yet clin-ically significant responses to treatment demonstrated by this modal-ity.49 The test-retest reproducibility of 3DE suggests a significantlearning curve that needs to be considered, and thus results cannotnecessarily be extrapolated to less experienced operators.Nevertheless, 3DE showed incremental benefit over conventional2D echocardiographic measures, suggesting an important role in as-sessing acquired RV pathology. Future work should focus onimproving spatial resolution to optimize RV endocardial delineation,in particular for adequate visualization of the RVOT in nondilatedright ventricles.

REFERENCES

1. de Groote P, Millaire A, Foucher-Hossein C, Nugue O, Marchandise X,Ducloux G, et al. Right ventricular ejection fraction is an independentpredictor of survival in patients withmoderate heart failure. J AmColl Car-diol 1998;32:948-54.

2. Ghio S, Gavazzi A, Campana C, Inserra C, Klersy C, Sebastiani R, et al. In-dependent and additive prognostic value of right ventricular systolic func-tion and pulmonary artery pressure in patients with chronic heart failure. JAm Coll Cardiol 2001;37:183-8.

3. Knauth AL, Gauvreau K, Powell AJ, Landzberg MJ, Walsh EP, Lock JE,et al. Ventricular size and function assessed by cardiac MRI predict majoradverse clinical outcomes late after tetralogy of Fallot repair. Heart 2008;94:211-6.

4. Sun JP, James KB, Yang XS, Solankhi N, Shah MS, Arheart KL, et al. Com-parison of mortality rates and progression of left ventricular dysfunction inpatients with idiopathic dilated cardiomyopathy and dilated versus nondi-lated right ventricular cavities. Am J Cardiol 1997;80:1583-7.

5. Ho SY, Nihoyannopoulos P. Anatomy, echocardiography, and normalright ventricular dimensions. Heart 2006;92(Suppl 1):i2-13.

6. Mertens LL, FriedbergMK. Imaging the right ventricle—current state of theart. Nat Rev Cardiol 2010;7:551-63.

7. Vonk-Noordegraaf A, Souza R. Cardiacmagnetic resonance imaging: whatcan it add to our knowledge of the right ventricle in pulmonary arterial hy-pertension? Am J Cardiol 2012;110:25S-31.

8. Shimada YJ, ShiotaM, Siegel RJ, Shiota T. Accuracy of right ventricular vol-umes and function determined by three-dimensional echocardiography incomparison with magnetic resonance imaging: a meta-analysis study. J AmSoc Echocardiogr 2010;23:943-53.

Journal of the American Society of EchocardiographyVolume 28 Number 3

Knight et al 373

9. Lang RM, Badano LP, TsangW, AdamsDH, Agricola E, Buck T, et al. EAE/ASE recommendations for image acquisition and display using three-dimensional echocardiography. J Am Soc Echocardiogr 2012;25:3-46.

10. Sugeng L, Mor-Avi V, Weinert L, Niel J, Ebner C, Steringer-Mascherbauer R, et al. Multimodality comparison of quantitative volu-metric analysis of the right ventricle. JACC Cardiovasc Imaging 2010;3:10-8.

11. Bhan A, Kapetanakis S, Monaghan MJ. Three-dimensional echocardiogra-phy. Heart 2010;96:153-63.

12. Galie N, Hoeper MM, Humbert M, Torbicki A, Vachiery JL, Barbera JA,et al. Guidelines for the diagnosis and treatment of pulmonary hyperten-sion: the Task Force for the Diagnosis and Treatment of Pulmonary Hyper-tension of the European Society of Cardiology (ESC) and the EuropeanRespiratory Society (ERS), endorsed by the International Society of Heartand Lung Transplantation (ISHLT). Eur Heart J 2009;30:2493-537.

13. Plockinger U, Gustafsson B, Ivan D, Szpak W, Davar J. ENETS consensusguidelines for the standards of care in neuroendocrine tumors: echocardi-ography. Neuroendocrinology 2009;90:190-3.

14. Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD,Chandrasekaran K, et al. Guidelines for the echocardiographic assessmentof the right heart in adults: a report from the American Society of Echocar-diography endorsed by the European Association of Echocardiography, aregistered branch of the European Society of Cardiology, and the Cana-dian Society of Echocardiography. J Am Soc Echocardiogr 2010;23:685-713. quiz 86–8.

15. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA,et al. Recommendations for chamber quantification: a report from theAmerican Society of Echocardiography’s Guidelines and Standards Com-mittee and the Chamber Quantification Writing Group, developed inconjunctionwith the EuropeanAssociation of Echocardiography, a branchof the European Society of Cardiology. J Am Soc Echocardiogr 2005;18:1440-63.

16. Howard LS, Grapsa J, Dawson D, Bellamy M, Chambers JB, Masani ND,et al. Echocardiographic assessment of pulmonary hypertension: standardoperating procedure. Eur Respir Rev 2012;21:239-48.

17. Hardegree EL, Sachdev A, Villarraga HR, Frantz RP, McGoon MD,Kushwaha SS, et al. Role of serial quantitative assessment of right ventric-ular function by strain in pulmonary arterial hypertension. Am J Cardiol2013;111:143-8.

18. Kjaergaard J, Petersen CL, Kjaer A, Schaadt BK, Oh JK, Hassager C. Eval-uation of right ventricular volume and function by 2D and 3D echocardi-ography compared to MRI. Eur J Echocardiogr 2006;7:430-8.

19. Ostenfeld E, CarlssonM, Shahgaldi K, Roijer A, Holm J. Manual correctionof semi-automatic three-dimensional echocardiography is needed for rightventricular assessment in adults; validation with cardiac magnetic reso-nance. Cardiovasc Ultrasound 2012;10:1.

20. Iriart X, Montaudon M, Lafitte S, Chabaneix J, Reant P, Balbach T, et al.Right ventricle three-dimensional echography in corrected tetralogy ofFallot: accuracy and variability. Eur J Echocardiogr 2009;10:784-92.

21. Niemann PS, Pinho L, Balbach T, Galuschky C, Blankenhagen M,Silberbach M, et al. Anatomically oriented right ventricular volume mea-surements with dynamic three-dimensional echocardiography validatedby 3-tesla magnetic resonance imaging. J Am Coll Cardiol 2007;50:1668-76.

22. van der Zwaan HB, Geleijnse ML, Soliman OI, McGhie JS, Wiegers-Groeneweg EJ, Helbing WA, et al. Test-retest variability of volumetric rightventricular measurements using real-time three-dimensional echocardiog-raphy. J Am Soc Echocardiogr 2011;24:671-9.

23. Puranik R, Tsang VT, Broadley A, Nordmeyer J, Lurz P, Muthialu N, et al.Functional outcomes after the Ross (pulmonary autograft) procedure as-sessed with magnetic resonance imaging and cardiopulmonary exercisetesting. Heart 2010;96:304-8.

24. Muthurangu V, Lurz P, Critchely JD, Deanfield JE, Taylor AM,Hansen MS. Real-time assessment of right and left ventricular volumesand function in patients with congenital heart disease by using high spatio-temporal resolution radial k-t SENSE. Radiology 2008;248:782-91.

25. Moledina S, Pandya B, Bartsota M, Mortensen KH, McMillan M,Quyam S, et al. Prognostic significance of cardiac magnetic resonance im-aging in children with pulmonary hypertension. Circ Cardiovasc Imaging2013.

26. Kind T, Mauritz GJ, Marcus JT, van de Veerdonk M, Westerhof N, Vonk-Noordegraaf A. Right ventricular ejection fraction is better reflected bytransverse rather than longitudinal wall motion in pulmonary hyperten-sion. J Cardiovasc Magn Reson 2010;12:35.

27. Steeden JA, Atkinson D, Hansen MS, Taylor AM, Muthurangu V. Rapidflow assessment of congenital heart disease with high-spatiotemporal-resolution gated spiral phase-contrast MR imaging. Radiology 2011;260:79-87.

28. Odille F, Steeden JA, Muthurangu V, Atkinson D. Automatic segmenta-tion propagation of the aorta in real-time phase contrast MRI usingnonrigid registration. J Magn Reson Imaging 2011;33:232-8.

29. Rosset A, Spadola L, Ratib O. OsiriX: an open-source software for navi-gating in multidimensional DICOM images. J Digit Imaging 2004;17:205-16.

30. Bland JM, Altman DG. Statistical methods for assessing agreement be-tween two methods of clinical measurement. Lancet 1986;1:307-10.

31. Tamborini G, Marsan NA, Gripari P, Maffessanti F, Brusoni D, Muratori M,et al. Reference values for right ventricular volumes and ejection fractionwith real-time three-dimensional echocardiography: evaluation in a largeseries of normal subjects. J Am Soc Echocardiogr 2010;23:109-15.

32. Pavlicek M, Wahl A, Rutz T, de Marchi SF, Hille R, Wustmann K, et al.Right ventricular systolic function assessment: rank of echocardiographicmethods vs. cardiac magnetic resonance imaging. Eur J Echocardiogr2011;12:871-80.

33. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas un-der two or more correlated receiver operating characteristic curves: anonparametric approach. Biometrics 1988;44:837-45.

34. van Wolferen SA, Marcus JT, Boonstra A, Marques KM, Bronzwaer JG,Spreeuwenberg MD, et al. Prognostic value of right ventricular mass, vol-ume, and function in idiopathic pulmonary arterial hypertension. EurHeart J 2007;28:1250-7.

35. Chin KM, Kim NH, Rubin LJ. The right ventricle in pulmonary hyperten-sion. Coron Artery Dis 2005;16:13-8.

36. van de Veerdonk MC, Kind T, Marcus JT, Mauritz GJ, Heymans MW,Bogaard HJ, et al. Progressive right ventricular dysfunction in patientswith pulmonary arterial hypertension responding to therapy. J Am CollCardiol 2011;58:2511-9.

37. Moller JE, Pellikka PA, Bernheim AM, Schaff HV, Rubin J, Connolly HM.Prognosis of carcinoid heart disease: analysis of 200 cases over two de-cades. Circulation 2005;112:3320-7.

38. van der Zwaan HB, Geleijnse ML, McGhie JS, Boersma E, Helbing WA,Meijboom FJ, et al. Right ventricular quantification in clinical practice:two-dimensional vs. three-dimensional echocardiography comparedwith cardiac magnetic resonance imaging. Eur J Echocardiogr 2011;12:656-64.

39. Tamborini G, Muratori M, Brusoni D, Celeste F, Maffessanti F, Caiani EG,et al. Is right ventricular systolic function reduced after cardiac surgery? Atwo- and three-dimensional echocardiographic study. Eur J Echocardiogr2009;10:630-4.

40. Shiran H, Zamanian RT, McConnell MV, Liang DH, Dash R, Heidary S,et al. Relationship between echocardiographic and magnetic resonancederivedmeasures of right ventricular size and function in patients with pul-monary hypertension. J Am Soc Echocardiogr 2014;27:405-12.

41. Grapsa J, O’Regan DP, Pavlopoulos H, Durighel G, Dawson D,Nihoyannopoulos P. Right ventricular remodelling in pulmonary arterial hy-pertension with three-dimensional echocardiography: comparison with car-diac magnetic resonance imaging. Eur J Echocardiogr 2010;11:64-73.

42. Morikawa T, Murata M, Okuda S, Tsuruta H, Iwanaga S, Murata M, et al.Quantitative analysis of right ventricular function in patients with pulmo-nary hypertension using three-dimensional echocardiography and atwo-dimensional summation method compared to magnetic resonanceimaging. Am J Cardiol 2011;107:484-9.

374 Knight et al Journal of the American Society of EchocardiographyMarch 2015

43. Tamborini G, Brusoni D, Torres Molina JE, Galli CA, Maltagliati A,Muratori M, et al. Feasibility of a new generation three-dimensional echo-cardiography for right ventricular volumetric and functional measure-ments. Am J Cardiol 2008;102:499-505.

44. Zhang QB, Sun JP, Gao RF, Lee AP, Feng YL, Liu XR, et al. Feasibility ofsingle-beat full-volume capture real-time three-dimensional echocardiog-raphy for quantification of right ventricular volume: validation by cardiacmagnetic resonance imaging. Int J Cardiol 2013;168:3991-5.

45. van der Zwaan HB, Helbing WA, McGhie JS, Geleijnse ML,Luijnenburg SE, Roos-Hesselink JW, et al. Clinical value of real-timethree-dimensional echocardiography for right ventricular quantificationin congenital heart disease: validation with cardiac magnetic resonanceimaging. J Am Soc Echocardiogr 2010;23:134-40.

46. Schattke S,WagnerM, Hattasch R, Schroeckh S, Durmus T, Schimke I, et al.Single beat 3D echocardiography for the assessment of right ventriculardimension and function after endurance exercise: intraindividual compari-son with magnetic resonance imaging. Cardiovasc Ultrasound 2012;10:6.

47. Fang F, Chan A, Lee AP, Sanderson JE, Kwong JS, Luo XX, et al. Varia-tion in right ventricular volumes assessment by real-time three-dimen-

sional echocardiography between dilated and normal right ventricle:comparison with cardiac magnetic resonance imaging. Int J Cardiol2013;168:4391-3.

48. Mor-Avi V, Jenkins C, Kuhl HP, Nesser HJ, Marwick T, Franke A, et al.Real-time 3-dimensional echocardiographic quantification of left ventric-ular volumes: multicenter study for validation with magnetic resonanceimaging and investigation of sources of error. JACC Cardiovasc Imaging2008;1:413-23.

49. vanWolferen SA, van de VeerdonkMC, Mauritz GJ, JacobsW, Marcus JT,Marques KM, et al. Clinically significant change in stroke volume in pul-monary hypertension. Chest 2011;139:1003-9.

50. Alfakih K, Plein S, Bloomer T, Jones T, Ridgway J, Sivananthan M. Com-parison of right ventricular volume measurements between axial andshort axis orientation using steady-state free precession magnetic reso-nance imaging. J Magn Reson Imaging 2003;18:25-32.

51. Fratz S, Schuhbaeck A, Buchner C, Busch R, Meierhofer C, Martinoff S,et al. Comparison of accuracy of axial slices versus short-axis slices formeasuring ventricular volumes by cardiac magnetic resonance in patientswith corrected tetralogy of fallot. Am J Cardiol 2009;103:1764-9.